The model-training companion for Morphosis

CUSTOM MODEL TRAINING

Create your own voice and sound models from existing recordings or dedicated studio sessions.

VOICE-FIRST POST-PRODUCTION

Built for dialogue, ADR and voice transformation workflows.

LOCAL AND OFFLINE

No cloud training. No external upload. Your models stay safe and controlled.

MADE FOR MORPHOSIS

Create models in BRAIN, then use them directly in MORPHOSIS.

Overview

BRAIN is a powerful model-training software designed for professional post-production teams.

It allows studios to create custom voice and sound models from existing audio archives or dedicated recording sessions, then use them directly in MORPHOSIS.

Running locally and offline, BRAIN helps teams keep control over their audio files, trained models and associated rights, with no cloud training and no external upload.

Designed primarily for voice-related post-production needs, BRAIN supports use cases such as voice continuity, ADR, altered or unavailable voices, age variations and alternative vocal versions.

As with any professional training tool, model quality depends directly on source quality: cleaner, better-prepared recordings lead to stronger models.

How BRAIN works

DRAG’N DROP YOUR OWN SAMPLES

Simply drag’n drop your own samples from your computer into BRAIN to access the preview and meta-data panel and customize your model.

PREVIEW AND META-DATA

After loading all the samples required to create your model, choose a model name and assign it to a category. The category will later appear in MORPHOSIS, making it easier to locate.

You can also add keywords to help you quickly find specific models in your library using the MORPHOSIS search bar.

START TRAINING

Once you’re done simply click the Start training button and let BRAIN do the work.

In Depth

Library Path

Click on the text box at the top of the interface to change the library path for the futur model.

Model Name

Click on the text box to name your model. This name will later appear inside MORPHOSIS under the USER tag

Category & Keywords

Choose a category name for your model, and a set of Keywords. Keywords make it easier to find your models in MORPHOSIS using the search function.

Data

DATA (Off / Light / Auto / Heavy) determines whether BRAIN should increase the size of your dataset by generating pitchshifted variations of your audio files. It can improve the overall quality of the resulting model but it will also increase processing and training time.

Model Queue

You can train multiple models consecutively. To add another model to the training queue, simply drag and drop a new audio file into this area.

Start Training

Start the training process for one or more models. Training time can vary depending on the performance of your computer and the size of the dataset and may take several hours to complete.

Create your audio models with BRAIN.

Use them in MORPHOSIS.

DISCOVER MORPHOSIS

The model-training companion for Morphosis

What is iLok?